"Voxel OX" 3D printer on OpenBuilds
Marshall Peck
Finally got around to submitting my 3D printer build to OpenBuilds.
Built a hand full of them over the last few months or so and they turned out pretty well.
Just thought I'd share.
critique/comment/enjoy :)
Protobuilds.com
mongoose.dog
> We've ompleted over half a dozen variations of this build
Missing a c in completed.
> A carpenter's square should be used to adjust your chop saw your saw.
Redundant "your saw"
Marshall Peck
Protobuilds.com
--
You received this message because you are subscribed to the Google Groups "ATXHS Discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to atxhs-discus...@googlegroups.com.
To post to this group, send email to atxhs-...@googlegroups.com.
Visit this group at http://groups.google.com/group/atxhs-discuss.
For more options, visit https://groups.google.com/d/optout.
Joseph Bridgewater
A few comments;
I feel like the top mounted Z axis motors would shift the center of gravity upwards, reducing the stability of the system. Maybe you have specific motivation(s) to have them top mounted?
In my experience with 'the' community at large, I've encountered people whom loath a moving Y axis bed to the extent they would never again using a moving Y axis bed, and advise others not to also. I don't recall ever encountering anybody that specifically wanted a moving Y axis bed. Maybe yours is significantly better than the ones they've encountered in the past.
The 20x80 beams sound very sturdy for CNC work. For less demanding work, such as a 3D printer, I wonder if adequate sturdiness could be achieved more economically?
Regarding flanged bearings as idlers: for every belt tooth that encounters the smooth round pulley surface, noise can be introduced into the work piece. I recommend using sprockets for idlers, instead of pulleys (or pulley-alike double flanged bearings).
Just for the sake of completeness; those 2x GT2 2mm 20T items listed on the BOM are not, in fact, "pulleys" as openbuilds (and many others) calls them. Because those items have teeth, but are not designed to mesh with other gears (ie; they're designed to mesh with belts), the correct term is: sprocket. I'd noticed openbuilds messed up even more on this product: http://openbuildspartstore.com/gt3-timing-belt-by-the-foot/ GT3 does not at all mean 3mm. In fact; 3mm pitch is not a size that real GT3 belts or sprockets are available in. The sort of relationship between GT2 and GT3 that openbuilds suggests does not exist at all. Same for this messed up product: http://openbuildspartstore.com/gt3-aluminum-timing-pulley-20-tooth/ Maybe they have other messed up products, IDK.
Regarding the use of 20T drive sprockets: I don't see any 16T (or even 18T) sprockets on openbuilds, but in the marketplace at large I've found 16T and 20T (and 18T) sprockets to have similar availability. May I suggest you work out the math of the carriage speed vs resolution for 16T (and 18T) sprockets instead of 20T and consider using 16T (or 18T) instead, even If they'd have to come from a different vendor, or perhaps show the speed differential vs resolution for the different sprocket tooth counts in a little chart. (BTW; I noticed openbuilds offers 14T ... I recommend against ever using 14T.)
"You can never have to many robots."
should be:
"You can never have too many robots."
Just FYI: I'd made a calculator for max carriage rate for the various electromechanical factors that play into that; for the specific steppers listed in your BOM, @ 12V motor voltage to the DRV8825s, with 20T 2mm drive sprockets, my calculator says your maximum axis rate is 359mm /sec (for 16T, it'd be: 287 mm/sec, and 18T would be 323mm /sec, with proportional improvements in resolution - a higher tooth count trades away resolution [all the time] for higher max speed).
Looking at LDO Motors' web site (and other research), I get the impression that maybe they no longer make the A variant of that motor, as depicted on openbuilds. The non A variant LDO-42TH47-1684 has 4.4 kg⋅cm holding torque instead of 5 kg⋅cm.
I'd recommend you physically orient your RAMPS board to maximize the potential for the DRV8825s to dissipate heat through convection. A large majority of the heat generated by the chips, by design, will be directed to the bottom of the chips, where they contact the little PCBs of the stepsticks. Without seeing it in person, I can only vaguely suggest: if you think of the sockets they're in as channels, then if you orient those channels vertically, it might present the best opportunity for air to flow upwards, through the channels. A properly placed fan would probably be a really good idea.
I find the openbuilds website to be absolutely maddening, as there's this persistent ugly grey block with some text demanding I sign up for something or other ALWAYS on the page, taking up a significant portion of the lower right corner. I have to use the browser's page composition examination features to disable it... I avoid web sites that do crap like that, if said crap isn't disabled by simply relegating them to the restricted security zone.
Riley
Marshal is getting fantastic prints day after day after day after day.... The moving bed is not a problem. A fixed bed means a gantry for the print head, which has it's own issues. (tanstaafl)
Smaller vslot works fine for a printer frame. The cost of larger vslot is minimal, looks great, and is rigid enough for light milling or huge multi-extruder print heads. (Awesome!)
For the scale these printers work at the electronics/pulleys have waaaaaay more resolution than the plastic extruders. Again, a non issue.
The open builds site works fine on my devices, Firefox, chrome, android, iPad. They have a page discussing actual pulley specs vs marketing/labeling and have chosen to go with what causes the least amount of customer confusion. :-\
-
R
This email was sent from my KLü Tablet
Marshall Peck
Thanks for check that stuff out. Appreciate the detail on the analysis. My replies below.
Top mounted motors
I top mounted the motors for a couple reasons; To keep them clean, space and to have the lead screw with a free end and not worry as much about backlash in the axial. Bottom mounted motors seem to collect a lot of dust and printer crud especially if I was going to do any sort of milling. I'm sure the bearings are sealed enough not to worry about it, but it bothers me enough to be on my list of reasons. Bottom mounting might have been a bit more tricky and a tight squeeze. There's always room up-top, so I went with that.
The couplings I'm using are sort of springy and this way, at least, they're pulled down against a bearing behind the mounting plate. The screw and motor shaft are pretty much touching nay, but a tiny bit. Another bearing captured on the other side and a lock nut would fix it. It'd need another plate with a pocket or at least clearance for the inner race. An of course, there's other coupling options too.
Sturdiness and 20x80mm economics
This machine is quite bottom heavy and rigid. 3D printing or knocking it around does't end up with any detectable flexing. I put it on it's side while printing once and it actually kept printing just fine save for a little motor noise on the X in the down direction :P Yes. VERY over built for a 3D printer, but I also had other plans such as light duty milling.
OpenBuilds extrusion is cheap. Especially the way I'm getting it now, which wasn't the way I was getting it at first + at the time, it kept me from having to purchase a bunch of different lengths (there's only 20x80 and 20x60 in the build). 20x80 is the only thing that fits the 8" build plate/carriage too. The back 2 wheels on the z-axis plates that are on the ends of the x-gantry could not have been tapped into the extrusion with the other profile sizes unless I used 3 wheels.
Flanged Bearing Idlers
On both X and Y axis the belts are actually twisted so that the back of the belt hits the bearing such that the teeth won't cause vibration. Not that I've really seen it an issue on other printers. Some Mendel90 builds and I'm sure others do it too. On all but the Y-motor mount, There's a way to angle the shaft of either the idler or the
I seen sprocket/pulleys for idlers. That'd work too but would require still 2 bearings holding the shaft and ultimately be more stuff and slight added cost.
Pulley vs Sprocket
I can see where you're coming from on that. And really, I have no idea. On Wikipeadia it seems pretty cut and dry, but on the ISO site things with teeth for belts seem to be referred to as pulleys. Didn't dig much further than that. For conformity's sake and search-ability, I'll probably keep it pulleys.
https://www.iso.org/obp/ui/#iso:std:iso:5294:ed-3:v1:en
Moving Bed Y-axis
I tend to agree with you there. I wanted it to make things more simple and cost effective. Main draw back is printing parts that aren't super stable like tall skinny things. When I encounter parts like that, I just print really slow. It's been an actual issue maybe 1 or 2 times. A 22in tall part on the more sloppily built 1m tall print I have had layers that looked the same as the lower layers. Other than that, the y-bed on this one is super solid. It's a bit more quick to build too. It can be a bit of a space hog and I wouldn't do it with a machine that has a build area more than 12". The 12" one I have now hangs off the edge of the table as it is.
Sprocket sizes and motor drives
0.0125mm x/y resolution is pretty good for me. Resolution concerns aside, it would be nice to have a bit more torque. Less teeth/smaller pulley diameter means more torque, but higher speeds on the stepper means less. I'm sure there's some point on those graphs where the 2 meet perfect. I'm curious why not 14 tooth?
Ok... let see about calculations. The pulley, ahem sprocket, is 12.2-ish mm wide.... so 38.3-ish circumference... annnnd if it goes 100mm.. er lets say 6000mm/minute (top speed I usually do travel moves) it's done 2.6 revolutions. So that's um... 156-ish rpm? There's a drop in torque doesn't seem to significant at that rpm (I only skimmed over some graphs, so please enlighten if anyone's looked into it further). Of course, drivers, acceleration and torque come in to play. On some I've got the drv8825 stepper drivers and others the A4988 drivers (I'm using RAMPS). It's still a crap shoot with manual current adjustment, but hey they're affordable.
One day I'll probably look into it if it becomes an issue... Any takers?
With 3D printing, step skipping is almost a non-occurrence. I imagine this is going to matter more when I'm continuously loading during milling operations, but I think I'll go back to all lead screw axis for that.
Curious about your calculatior. FYI, I'm using these steppers (76oz-in). I thought the top rpm @12v was decently high like in the 600 range at least.
I just upped the top speed limits to one sitting next to me to 1000mm/sec (default and max) and 8000 mm/sec² acceleration (default and max). Pronter face lets me jog it 300mm and it seems to get up there at 25000mm/min/400mm/sec.
And if the graph at the bottom is correct, I may have just seen my machine get up to like 6 or 700mm/sec. maaaay have skipped on the decel or it was the motors powering off.
All my driver heat sinks are all randomly oriented. A lot of them didn't start out on these machines anyway. Hans put a nice RAMPS mount with fan on his printer. It blows on em' pretty good.
Too many robots
I searched the build page, comments... I vaguely recall typing it, but for the life of me can't find where it is. Please help me find the typo.
Protobuilds.com
Marshall Peck
Maybe he'll get in here in a bit and lay down some mo 3D printer knowledge.